Inaccurate orf length in gff3 file

[mollylRivers]

Hi Brian,

I am using TransDecoder on my de novo assembled transcriptome, which has been assembled using illumina short-reads and iso-seq long-reads in rnaSPAdes.

Script: TransDecoder.LongOrfs -t assembled.Transcriptome.fasta -m 100, TransDecoder.Predict -t assembled.Transcriptome.fasta --single_best_only

When using the gff3 output file from TransDecoder I am finding an issue with the annotated length of the selected open reading frame. There is a discrepancy between the annotated sequence length (in the gff3 file) and the actual sequence length. The sequence is labelled as being longer than the actual contig (which I have manually checked using the longest.orfs.cds file). What I am finding, is that the selected orf is being annotated with the length of the longest orf, even when this is not the one that was selected.

I am not sure if this has to do with the iso-seq long-read sequences in the transcriptome, but it is causing issues with another programme I need to run with the TransDecoder gff3 output file. Any input you can provide on what might be causing this issue and how to fix this would be greatly appreciated.

Many thanks,
Molly Rivers

[brianjohnhaas]

This sounds pretty peculiar. If you can send me the example, I'd be happy to take a look at it. bhaas at broadinstitute dot org best, B

…

On Wed, Nov 1, 2023 at 9:53 AM mollylRivers ***@***.***> wrote: Hi Brian, I am using TransDecoder on my de novo assembled transcriptome, which has been assembled using illumina short-reads and iso-seq long-reads in rnaSPAdes. Script: TransDecoder.LongOrfs -t assembled.Transcriptome.fasta -m 100, TransDecoder.Predict -t assembled.Transcriptome.fasta --single_best_only When using the gff3 output file from TransDecoder I am finding an issue with the annotated length of the selected open reading frame. There is a discrepancy between the annotated sequence length (in the gff3 file) and the actual sequence length. The sequence is labelled as being longer than the actual contig (which I have manually checked using the longest.orfs.cds file). What I am finding, is that the selected orf is being annotated with the length of the longest orf, even when this is not the one that was selected. I am not sure if this has to do with the iso-seq long-read sequences in the transcriptome, but it is causing issues with another programme I need to run with the TransDecoder gff3 output file. Any input you can provide on what might be causing this issue and how to fix this would be greatly appreciated. Many thanks, Molly Rivers — Reply to this email directly, view it on GitHub <#190>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABZRKXZTVLPPYSHDECTHV43YCJH57AVCNFSM6AAAAAA6ZKKAF2VHI2DSMVQWIX3LMV43ASLTON2WKOZRHE3TENBSGU4DONY> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- -- Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas <http://broad.mit.edu/~bhaas>

[mollylRivers]

Hi Brian,

Thanks so much for your quick response. I have attached the transdecoder produced gff3 file, a sample of the longest_orfs cds file (it was far too big for me to attach the whole thing), along with a txt file showing the error message produced by cellranger which explains the discrepancy in read length I was explaining.

Many thanks,
Molly

transdecoder_produced_gff3_file.gff3.gz
cellranger_error_message.txt.gz
longest_orfs.txt

[brianjohnhaas]

Hi Molly, I think the issue here has to do with the expected formatting, as cellranger wants a gtf rather than a gff3 file. You can try converting the gff3 file you have to gtf format using the script included with TransDecoder like so: TransDecoder/util/gff3_gene_to_gtf_format.pl Trinity.fasta.transdecoder.gff3 Trinity.fasta. > transdecoder.gtf and then try running cellranger with the transdecoder.gtf file Let's see if that works. best, B

On Thu, Nov 2, 2023 at 5:45 AM mollylRivers ***@***.***> wrote: Hi Brian, Thanks so much for your quick response. I have attached the transdecoder

produced gff3 file, a sample of the longest_orfs cds file (it was far too big for me to attach the whole thing), along with a txt file showing the error message produced by cellranger which explains the discrepancy in read length I was explaining.

Many thanks, Molly transdecoder_produced_gff3_file.gff3.gz cellranger_error_message.txt.gz longest_orfs.txt — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID:

***@***.***>

…

-- -- Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas

[mollylRivers]

Hi Brian,

Sorry, I should have said that I have already reformatted the gff3 file to a gtf, and used this file for cellranger. The error message I attached, is produced when using the gtf file.

Many thanks,
Molly

[brianjohnhaas]

I see. When running cellranger, are you giving the original Trinity.fasta file as input along with that gtf file? It's complaining about the sequence length, but the sequence length for that Trinity contig should be ~1.5kb, which can be verified from the fasta file, right? The longest_orfs file will have the shorter one corresponding to just the cds sequence, and that wouldn't match up with the gtf or gff3. best, B

…

On Thu, Nov 2, 2023 at 9:50 AM mollylRivers ***@***.***> wrote: Hi Brian, Sorry, I should have said that I have already reformatted the gff3 file to a gtf, and used this file for cellranger. The error message I attached, is produced when using the gtf file. Many thanks, Molly — Reply to this email directly, view it on GitHub <#190 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABZRKX7T7CWV5DYEHQYWGQTYCOQK5AVCNFSM6AAAAAA6ZKKAF2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOJQG43TMNZVGE> . You are receiving this because you commented.Message ID: ***@***.***>

-- -- Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas <http://broad.mit.edu/~bhaas>

[mollylRivers]

Hi Brian, thanks so much for the information. We were using the TransDecoded transcriptome for cellranger, and this does in fact seem to be the issue. Thanks you for your help :)
Molly